Comparative Energy Transition Policy: How Exposure, Policy Vulnerability and Trust Affect Popular Acceptance of Policy Expansion

Abstract

This article examines how exposure to energy transition and climate policy vulnerability influence popular support for more ambitious climate policy. Moreover, it explores whether this relationship depends on a person’s generalized and political trust. Comparing data from surveys in Germany and Switzerland, the findings reveal that perceived exposure to energy transition positively influences climate policy support, while individual climate policy vulnerability decreases it. For individuals with higher levels of trust, exposure helps enhance the positive effect (subjective exposure) or dampen the negative effect (policy vulnerability). These results underscore the importance of incorporating trust and subjective perceptions into climate policy frameworks.

Keywords:

• climate policy
• comparative
• policy exposure
• subjective exposure
• political trust
• generalized trust

Introduction

Energy transition in Europe needs to proceed at an accelerated pace due to both the evolving climate crisis and new geopolitical challenges. European countries are under pressure to find policy solutions to phase out fossil fuels and replace them with clean renewable energy. In strengthening existing policy (e.g. speeding up permit procedures for renewable energy, increasing climate investment, increasing carbon prices, ramping up the energy efficiency of buildings, decarbonizing transportation), policymakers need to implement effective solutions while sustaining public support for more ambitious climate and energy policy. This paper investigates how individual exposure, policy vulnerability and trust contribute to shaping public endorsement for the expansion of such policies.

To conceptualize policy expansion, we examine attitudes towards an increase in (or introduction of) national carbon taxes. Carbon taxes are a prominent policy proposal to steer people away from socially and environmentally costly fossil fuels and incentivize firms and individuals to switch to greener energy sources. However, actual carbon prices have consistently fallen short even of conservative estimates for the “social cost of carbon” (Green 2021), and hence increases in existing carbon taxes are certainly needed to attain relevant emission reductions. Moreover, carbon taxes are generally unpopular, and the distributional effects of carbon taxes have been found to decrease popular acceptance drastically (Beiser-McGrath and Busemeyer 2023; Schaffer 2023).

We first argue that perceptions of exposure to energy transition and considerations about individual policy vulnerability matter for attitudes towards more ambitious climate policy (e.g. an increase in carbon taxes). In terms of exposure, past decades have brought about considerable transformation in the energy sector and have elevated the level of renewable energy generation. Nations developed different schemes to incentivize renewable energy generation (Schaffer and Bernauer 2014; Bayer and Urpelainen 2016) and considerably increased renewable energy generation. The accompanying decentralization of energy supply with wind turbines and solar parks made the energy transition more visible to the public. However, people’s perceptions of this exposure to energy transition via visible infrastructure vary depending on where they live.

Regarding policy vulnerability to climate policy expansion, we submit that people’s individual climate policy vulnerability varies, according to their reliance on fossil fuels, for example. Overall, we argue that the more exposed people feel and the more vulnerable they are to the policy change, the more negative they are towards leveling up climate efforts.

In a second step we then explore whether levels of generalized and political trust can ameliorate the effect of exposure on support for policy expansion. Are individuals with high trust more accommodating of governmental proposals to ratchet up policy even when they are highly exposed or vulnerable to these policies? With respect to the trust–policy support nexus, previous studies have found that people with higher trust levels have a higher level of support for carbon taxes. Hence, those with higher trust levels in the political process believe that new policies can bring about societal benefits even if these policies personally disadvantage them (Fairbrother 2019; Umit and Schaffer 2020). While this direct effect of trust on climate policy support is well documented in the literature, the novelty of our approach is that we investigate how trust affects the relationship between exposure, vulnerability and support for policy expansion. We propose that people who are exposed to energy transition and/or vulnerable to the policy change but exhibit high generalized or political trust levels will support policy expansion.

Lastly, we explore whether people in countries that have been exposed to fewer energy transition policies differ in their attitudes towards new policies. Hence, our comparative policy analysis scrutinizes how contextual factors affect the relationships mentioned above. In our empirical section, we leverage nationally representative survey data from Switzerland and Germany for our principal analysis and compare these to survey data from Hungary, an eastern European EU country with a comparatively shorter history of climate policy action.

Our results show that the two exposure measures (exposure to energy transition and policy vulnerability) have different implications for individual support for the expansion of climate policy. Subjective exposure to (past) energy transition is positively related to a person’s support for climate policy expansion. This effect, however, is mainly driven by people’s perceived exposure to solar energy. Moreover, both generalized and political trust significantly increase the positive relationship between exposure and policy support. Whereas for low-trust individuals higher perceived exposure leads to less support for climate policy expansion, high-trust individuals with a high subjective exposure to renewable energy infrastructure show the highest levels of support.

To assess how people might be affected by climate policy expansion (policy vulnerability), we look at factors such as how often they use their cars, whether they own or rent their homes, and the age of the buildings they live in. Interestingly, we observe a significant and negative association of these factors with the support for climate policies, which indicates that individuals tend to evaluate proposals based on their personal policy vulnerabilities. However, even if a person is affected by or exposed to climate change policies, we discovered that higher levels of social and political trust could help lessen the negative connection between being directly affected by policies and supporting more ambitious climate measures.

We contribute to the rapidly expanding literature on individual attitudes towards climate policy in at least three important ways. The first contribution lies in explicitly considering policy expansion – that is, our main outcome variable asks for people’s opinion on climate policy expansion. In order to implement more stringent climate and energy policies towards the societal goal of energy transition, we need to know more about people’s support for increasingly ambitious and effective climate policy. Second, to the best of our knowledge, we are the first to explicitly consider measures of subjective exposure to energy transition and their impact on political attitudes. Our argument is that we need to go beyond the standard demographic variables and consider how existing infrastructural changes to people’s communities and local areas may have affected their support for more ambitious climate policy. Third, adding to a broad literature that has consistently pointed to the importance of trust for individual (climate policy) attitudes (Hammar and Jagers Sverker 2006; Delhey et al. 2011; Fairbrother 2019; Fairbrother et al. 2019; Davidovic and Harring 2020; Umit and Schaffer 2020), we develop and test arguments about how individual trust conditions the (proposed negative) relationship between exposure and support for climate policy expansion.

State of the Art and Argument

Impact of Exposure and Vulnerability on Support for Policy Expansion

To construct our argument on how exposure and vulnerability affect public policy support, we introduce two concepts that we consider crucial for an individual’s support for policy expansion: (1) individuals’ perceptions of exposure to energy transition and (2) individual vulnerability towards more ambitious policy expansions.

In line with the literature on policy feedback, we argue that existing policies shape present politics and policymaking (Lowi 1964; Pierson 1993; Béland 2010). Our interest pertains to climate-related energy transition policies designed to completely transform fossil-fuel-dependent economies to rely on renewables instead. To reach this goal, countries have taken different paths and implemented different policy mixes (Kern et al. 2019). Most industrialized countries have incentivized renewable energy production over the past 20–30 years – for example, via feed-in tariffs or quota systems. While research shows that these renewable energy efforts successfully reduced the respective national carbon emissions (Le Quéré et al. 2019), locally they also led to greater visibility of renewable energy infrastructure, such as wind turbines and solar panels. We highlight two possible pathways for how exposure to renewable energy infrastructure can be linked to people’s preferences for more ambitious policies.¹

First, individuals, households and broader communities have been able to actively participate in this transition, for example by mounting solar panels on their roofs or balconies. They have hence obtained potentially valuable information about individual (positive and negative) implications of the large societal project of energy transition.² On the one hand, people might have experienced positive implications from mounting solar panels themselves and might understand how their energy costs drop as a consequence and how they would be less affected by a higher carbon tax. On the other hand, however, having renewable energy infrastructure within one’s vicinity might be associated with negative visual or audio impact (Hirsh and Sovacool 2013), place attachment (Devine-Wright 2009), adverse environmental impacts (Ansolabehere and Konisky 2014) or lower property value (Petrova 2013; Cashmore et al. 2019). Wind turbines are very visible in some rural regions that are better suited to wind power generation.³ Overall, regional variation in renewable energy potential, investment and implementation have led to differences in individual exposure. Due to directly observing the positive or negative impacts of these changes, the net effect on people in regions with more visible infrastructure remains unclear.

Second, following up on the potentially negative personal implications from exposure portrayed above, we contend that the ongoing politicization of both renewable energy infrastructures and ambitious climate policies (such as carbon taxes) could explain why an individual’s exposure to energy transition and their overall stance on climate policy expansion are linked. Notably, right-wing populist parties strategically position themselves against both energy transition infrastructure and broader, more ambitious climate policies (Otteni and Weisskircher 2022; Schaffer 2023), and thus influence how people connect their exposures, policy vulnerabilities and attitudes to more general climate policy progression.

The second concept that we consider relevant to determine policy support is individual policy vulnerability. The implementation of more ambitious climate and energy policies to transition away from fossil-fuel-based economies has distributional consequences (Aklin and Mildenberger 2020; Colgan et al. 2021; Gaikwad et al. 2022). In our context, policy vulnerability means climate change policies will be costlier for some than for others. We focus on the fact that carbon pricing makes fossil-fuel-based activities more expensive than alternatives. This disproportionately affects individuals who drive more often and those who, for example, depend on older fossil-fuel heating systems. As carbon-intensive activities become more expensive with higher carbon taxes, we expect that people who are more exposed to the policy consequences (who have a higher vulnerability) will be less supportive of policy expansion. This definition of policy vulnerability centers on an individual’s general vulnerability profile but does not account for an individual’s employment-related vulnerabilities (Gaikwad et al. 2022). Several studies find empirical evidence for self-interested responses to the higher costs associated with carbon pricing (Carattini et al. 2018; Douenne and Fabre 2020). Further, these considerations have been found to shape general climate and energy policy preferences (Drews et al. 2016; Bergquist et al. 2020).

Overall, there are convincing arguments on how subjective exposure to renewable energy policy and individual policy vulnerability can negatively impact an individual’s finances, place attachment or standard of living. Thus, we argue that these types of exposures may lead to less support for a more ambitious climate policy.

H1: People with higher exposure and vulnerability to energy transition are less supportive of climate policy expansion.

How Trust Influences the Link between Exposure, Policy Vulnerability and Support for Policy Expansion

Past contributions have found a robust link between measures of individual trust and individual climate policy support (Hammar and Jagers Sverker 2006; Fairbrother et al. 2019; Davidovic and Harring 2020). Even more relevant for the focus of this paper, empirical studies have shown that climate policy ambition increases with the level of trust (Rafaty 2018; Douenne and Fabre 2020). The literature usually refers to political and social (generalized) trust.

More specifically, in the literature on the acceptance of carbon taxes, political trust has been found to be important to explain policy support (Cologna and Siegrist 2020; Umit and Schaffer 2020; Bergquist et al. 2022; Fairbrother 2022). One of the greatest hindrances to carbon tax support is skepticism about whether the government will use the extra tax money to combat climate change or to supplement general government spending (Klok et al. 2006; Hsu et al. 2008). If people believe political actors are genuinely doing their best for society, they will support political decisions (such as the decarbonization of the economy and more ambitious climate policy) and will be more willing to accept detrimental outcomes even if these affect them personally. Hence, we argue that when people trust in the long-term direction of governmental policy, they will support policy expansion more, even if they have been – or will be – exposed or vulnerable to the consequences of the policy.

Similarly, a broader form of trust, generalized trust, has been positively linked to people’s carbon tax preferences (Davidovic and Harring 2020), their climate behavior (Tam and Chan 2018) and their willingness to pay (Smith and Mayer 2018; Cologna and Siegrist 2020). Thus, high generalized trust may help to solve collective action problems such as climate change (Nannestad 2008; Smith and Mayer 2018). This diffuse form of trust refers to expectations about other people’s preferences: it describes a view that people would not intentionally harm others if they could avoid it (Delhey et al. 2011). Smith and Mayer (2018, p. 141) argue that trusting others to do the same is key for people to change their behavior. With respect to climate policy support, Davidovic and Harring (2020, p. 3) point out: “Generalized trust is linked to policy support since people also need to trust their fellow citizens and other actors to comply with the policies in order to accept their implementation.” Analogously, we believe that generalized trust in other citizens will ameliorate the potentially negative effects of exposure to energy transition as individuals believe that others also contribute to the shared societal goal.

We argue that trust can ameliorate the negative relationship (H1) between exposure, policy vulnerability and policy expansion. Hence, people who are exposed and/or vulnerable to energy transition but have a high generalized or political trust will feel less negative about a policy expansion.

H2: Trust positively affects the relationship between exposure, policy vulnerability and policy expansion.

Data and Design

Survey

Our analysis is based on data from nationally representative surveys in Germany, Switzerland and Hungary. In Germany, we conducted our survey just before the September 2021 elections; the Swiss survey was fielded in the summer of 2022, one year after the electorate rejected a proposal for a more ambitious CO₂ law, and the Hungarian survey was fielded before the general elections in April 2022. Sonda Ipsos recruited an online representative sample of 3,500 subjects from Germany and 2,500 each from Switzerland and Hungary.⁴

We selected Germany and Switzerland for our principal analysis for several reasons. First, these two countries boast a long history of climate and energy transition policies. In Germany, significant funding for the Erneuerbare-Energien-Gesetz (Renewable Energy Sources Act) (EEG) since the year 2000 marks the inception of the energy transition (“Energiewende”). Similarly, in Switzerland, the enactment of the CO₂ law in 1999, the subsequent introduction of the CO₂ levy in 2008, and, finally, the 2017 decision on Energy Strategy 2050 constitute significant policy innovations aimed at achieving carbon neutrality in Swiss energy production. Therefore, policymakers in Germany and Switzerland have addressed various aspects of climate and energy policies and have already implemented far-reaching measures. Second, voters in both countries have been “exposed” to energy transition policies. However, regional differences in (past) exposure exist in both countries, depending on factors such as regional opportunities for renewable energy generation. Additionally, disparities exist in the vulnerability or exposure of individuals to carbon pricing policies.

Hungary, being economically less developed than Germany and Switzerland, only commenced its energy transition upon accession to the EU. The country has consistently ranked in the bottom third of EU nations in terms of energy transition, with decarbonization becoming a primary focus of its energy strategy as recently as 2020 (Hungary, Department of Innovation and Technology 2020). Although discussions regarding the implementation of a carbon tax have surfaced, no such measure has been enacted to date. By including Hungary in our selection of countries, we aim to explore similarities and differences in exposure to energy transition policies and how these relate to support for climate policy. Thus, after testing our hypotheses on two frontrunners in energy transition (Switzerland and Germany), we validate our findings by examining whether the overall results also apply to Hungary, where exposure has been comparatively limited.

Operationalization

Dependent Variable: Policy Expansion

To operationalize our concept of a policy expansion towards a more ambitious climate policy, we used a question on a potential carbon tax increase. We asked respondents in Germany and Switzerland whether and to what extent they would support a CO₂ tax increase if this were necessary.⁵ Respondents could choose an answer on a scale from 1 (support) to 4 (do not support) or choose a “don’t know” option. Our dependent variable was whether or not people would support an increase in the CO₂ tax (Germany, Switzerland). To make the results more intuitive, we used a reversed scale for the statistical analysis: 4 meant that the respondent would support a carbon tax increase. To better visualize the regional variation of the dependent variable, online Appendix Figure A1a shows average values of support for a CO₂ tax increase in the 16 German Länder. Support is especially low in the eastern part of the country and is highest in the city states and the more populous states including North Rhine-Westphalia and Baden-Württemberg.

We used a similar measurement in Hungary, where people could express their support for the introduction of a carbon tax on a 1 (strongly support) to 5 (strongly oppose) scale (or could choose “don’t know” instead).⁶

Independent Variables: Exposure and Trust

The focal independent variables are exposure to renewable energy sources (subjective exposure), exposure to policy expansion (policy vulnerability), as well as trust (generalized and political trust).

In order to operationalize individual exposure to energy transition and policy vulnerability, we leveraged two different measures. First and most innovatively, we asked our respondents different questions about energy transition in their local area. This is mostly based upon exposure to visible energy infrastructure and thus is a measure of energy transition exposure. The questions used pertain only to solar and wind power. However, as these are the renewable technologies that have exhibited the largest increase in capacity (IEA) and are readily observable and sometimes controversial (Carley et al. 2020), we are confident that they effectively capture subjective exposure. To approximate individual exposure to wind energy infrastructure, we first asked respondents whether the statements “there are one or more wind turbines in my local area/commune” or “I see one or more wind turbines every day” applied to them or not. To account for solar exposure, we presented three statements: “There are solar panels in my community”, “At least one of my acquaintances has solar panels” and “I have solar panels on my roof”. These different statements proxy the different levels of exposure from merely seeing solar panels to owning them and actively participating in the process of energy transition.

Figure 1 shows the distribution of answers to these exposure questions for Germany and Switzerland. As expected, subjective exposure to wind energy is much lower in Switzerland than in Germany. In Switzerland, there are only a handful of wind farms and some are in remote mountainous terrain (Umit and Schaffer 2022). Also, wind energy production per capita is around 17 kwh compared to 1,619 kwh in Germany (Hälg and Nipkow 2021). With respect to solar energy, subjective exposure is more equal, although there is still a marked difference, Germany has higher per capita production (614 kwh) compared to Switzerland (297 kwh).⁷

Figure 1. Distribution of subjective exposure to energy transition in Switzerland and Germany (self-reported, own survey)

Second, we accounted for respondents’ policy vulnerability by approximating their individual policy vulnerability to more ambitious climate policy. To this end, we constructed an index and measured how much respondents would be exposed to the specific policy expansion at hand: an increase of the CO₂ tax. In the survey, we asked respondents how frequently they used their cars (1–5), whether they owned their house/flat or were tenants (0,1)⁸ and, finally, whether they lived in an old house (“Altbau”) or a new one (“Neubau”) (1,0). The minimum value of this policy vulnerability index was 0 (for someone who is a homeowner, lives in a new house and does not use their car) while the maximum was 8 (for someone who rents their apartment/house, lives in an old house and uses their car daily).

We measured individual trust using two different metrics: generalized trust and political trust. While political trust has robustly been found to influence individual climate policy attitudes (Hammar and Jagers Sverker 2006; Fairbrother 2019; Fairbrother et al. 2019; Davidovic and Harring 2020; Umit and Schaffer 2020), generalized trust has been a less prominent determinant of people’s climate policy preferences (Cologna and Siegrist 2020; Bergquist et al. 2022), but a significant predictor of people’s carbon tax preferences (Davidovic and Harring 2020). As political and social (generalized) trust are related, we tested both in different models (Uslaner 2018). Generalized trust is measured by a well-established question asking people to evaluate on a scale from 1 to 5 whether they believe that you cannot be too careful (1: Man kann nicht vorsichtig genug sein) or most people can be trusted (5: Man kann den meisten Menschen trauen). This measure has been found to be a reliable and valid way to measure general(ized) trust (Delhey et al. 2011).

Political trust was operationalized using two different items from our survey: We constructed an index of political trust by averaging how much respondents indicated they trusted the government and parliament. First, we asked people about their trust in several different institutions on a 0–10 scale where 0 corresponded to “do not trust at all” and 10 to total trust (the government, the media, the EU, parliament). Then we averaged their trust level towards the two domestic institutions: the government and parliament. We did this to account for a possible aversion to the current government by the supporters of the opposition. All these items have been used and applied widely in previous empirical research (e.g. Kulin and Sevä 2021).⁹

Control Variables

Our control variables were the self-placement of people on the left–right ideological scale (Ideology, where 1 means Left and 10 means Right); concern about climate change (from 1 “not worried” to 4 “very worried”); interest in politics (from 1 “not interested” to 4 “very interested”), age group (below 25, 25–50, 50–75, above 75), gender (1 men, 2 women), and income (below 1Q of the income distribution, between 1Q and median, between median and 3Q, above 3Q). We expected that people who were more left-wing, would be more concerned about climate change, and younger would be more supportive of energy transition in general and of ambitious climate change policy more specifically (Bergquist et al. 2022).

Modeling Strategy

First, we use a mixed-effect linear model to account for the hierarchical nature of our data-generating process, which allows for both fixed and random effects across the regions within the countries to be taken into account.¹⁰ Second, we model binary policy support (as a yes or no) with logit models within the individual countries to show how the results vary below the hierarchical structure and how the respondents behave in each of the countries.

Results

In our theory section, we hypothesized that people who felt exposed to energy transition policies (both via past policies and vulnerability to current policies) might support a possible increase of the CO₂ tax less than others, all else being equal. Moreover, we argued that generalized and political trust can reduce the negative effects of exposure on policy support.

Table 1 illustrates how our exposure measures are correlated with support for a carbon tax increase (M1 and M5) and how generalized and political trust impact these relationships (M2–4; M6–8). In the main analysis, we only use the German and Swiss survey data.¹¹

Table 1. Marginal influence of subjective and policy vulnerability to energy transition on the support of carbon tax increase for generalized (M1–M4) and political (M5–M8) trust (mixed effect linear models with regions).

	Generalized trust				Political trust
DV: Support for CO2 tax increase	M1	M2	M3	M4	M5	M6	M7	M8
Subjective exposure	0.29 ***		−0.41 **	0.29 ***	0.29 ***		−0.03	0.29 ***
	(0.06)		(0.15)	(0.06)	(0.06)		(0.12)	(0.06)
Policy vulnerability	−0.10 ***	−0.11 ***	−0.10 ***	−0.16 ***	−0.10 ***	−0.10 ***	−0.10 ***	−0.14 ***
	(0.01)	(0.01)	(0.01)	(0.02)	(0.01)	(0.01)	(0.01)	(0.02)
Generalized trust	0.17 ***	0.17 ***	0.07 **	0.07
	(0.01)	(0.01)	(0.02)	(0.04)
Political trust					0.09 **	0.09 ***	0.07 ***	0.05 **
					(0.01)	(0.01)	(0.01)	(0.02)
Ideology	−0.04 ***	−0.04 ***	−0.05 ***	−0.04 ***	−0.04 ***	−0.04 ***	−0.04 ***	−0.04 ***
	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
Political interest	0.08 ***	0.08 ***	0.08 ***	0.08 ***	0.08 ***	0.08 ***	0.08 ***	0.08 ***
	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)
Climate concern	0.47 ***	0.47 ***	0.47 ***	0.47 ***	0.42 ***	0.42 ***	0.42 ***	0.42 ***
	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)	(0.02)
Income	0.00	−0.00	0.00	0.00	−0.01	−0.01	−0.01	−0.01
	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
Age	−0.01 ***	−0.01 ***	−0.01 ***	−0.01 ***	−0.01 ***	−0.01 ***	−0.01 ***	−0.01 ***
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Gender	−0.02	−0.02	−0.02	−0.02	−0.01	−0.01	−0.01	−0.01
	(0.03)	(0.03)	(0.03)	(0.03)	(0.03)	(0.03)	(0.03)	(0.03)
Subjective exposure (solar)		0.27 ***				0.28 ***
		(0.05)				(0.05)
Subjective exposure (wind)		0.03				0.03
		(0.04)				(0.04)
Subjective exposure*Generalized trust			0.24 ***
			(0.05)
Policy vulnerability*Generalized trust				0.02 **
				(0.01)
Subjective exposure*Political trust							0.06 **
							(0.02)
Policy vulnerability*Political trust								0.01 **
								(0.00)
(Intercept)	1.14 ***	1.13 ***	1.45 ***	1.42 ***	1.33***	1.32 ***	1.47 ***	1.56 ***
	(0.11)	(0.11)	(0.13)	(0.16)	(0.11)	(0.11)	(0.12)	(0.14)
Num. obs.	5050	5050	5050	5050	5034	5034	5034	5034
Num. groups: Bun_Cou_Can	42	42	42	42	42	42	42	42

* p < 0.05, ** p < 0.01, *** p < 0.001

We only partially confirm our hypotheses. Model 1 shows that – contrary to our expectations – if a respondent’s self-reported exposure to renewable energy infrastructure changes from no exposure (0) to high exposure to renewable energy sources in the neighborhood and at home (1) (all else being equal), we predict an 8 per cent increase in the respondent’s support for a carbon tax increase. Although the effect size is not large, it is statistically significant by conventional standards. Hence a higher exposure to infrastructure is linked to a more supportive stance on ambitious climate policy.

To account for the potentially different effects due to exposure to wind turbines compared to solar panels, we split the subjective exposure index and replace it with separate wind and solar indices in Model 2. People might consider solar panels less intrusive than wind turbines, which alter the view of the landscape (Umit and Schaffer 2022). Indeed, we find that exposure to solar panels is associated with a 6 per cent increase in support for the increase in a national carbon tax, while there is no correlation between exposure to wind turbines and support for carbon tax increases.¹² There might be a concern that heightened politicization and polarization of the issue may lead individuals who are less satisfied with the policy to also perceive themselves as more exposed to renewable energy, potentially influencing our results. We cannot completely rule out this possibility since our survey questions were asked simultaneously, but robustness checks using more objective measures of exposure do not change our substantive results.¹³

The potential individual policy vulnerability, however, is associated with greater opposition to climate policy expansion, in line with our Hypothesis 1. Model 1 shows that a unit increase in policy vulnerability, defined as high carbon usage for heating or transportation, decreases support for a more stringent carbon tax by 11 per cent.

Moreover, in line with previous literature (Cologna and Siegrist 2020; Davidovic and Harring 2020; Fairbrother 2022), our results show that both generalized trust and political trust positively affect support for an increase in carbon taxes. As Model 1 in Table 1 illustrates, a unit increase in generalized trust is associated with a substantive 17 per cent increase in support for policy expansion.

In Models 3 and 7 we then explore how a person’s generalized (M3) and political (M7) trust interact with their subjective exposure to renewable energy sources. To better interpret our results, we plot the marginal effects in the upper panel of Figure 2. Figure 2a shows that if the respondent has very little general trust in other people (a score of 1 on a 5-point scale), higher levels of subjective exposure (on the x-axis) make them less supportive of future climate policy expansion (support of a carbon tax increase on the y-axis). We predict that for a person who trusts others the least, an increase in renewable energy exposure from the minimum to the maximum (being exposed in the city, in the neighborhood and at home) would decrease their support for a carbon tax increase from 1.8 to 1.5 (a 7.5 per cent change). This negative effect – that we had initially proposed in our H1 – however, vanishes and turns positive the more trusting the person is. We predict that a person who trusts most people (5) would increase their support for a carbon tax increase from 2.5 to 3 (a 12.5 per cent change) if exposed to past energy transition (1) compared to no exposure at all (0). The results we obtain from our political trust measure (Figure 2b) are very similar but less stark. Also here, political trust clearly has a mitigating impact on how exposure affects climate policy support.

Figure 2. Marginal influence of subjective exposure (a, b) and policy vulnerability (c, d) on support for a carbon tax increase at varying levels of trust (Germany and Switzerland)

Models 4 and 8 demonstrate how both generalized and political trust can impact the (negative) correlation observed between an individual’s policy vulnerability and their support for policy expansion. Figure 2c illustrates that higher levels of generalized trust mitigate the negative effect of the policy vulnerability variable. Specifically, we anticipate that individuals with the lowest levels of trust would decrease their support from 2.38 to 1.5 (a 22 per cent change) when highly exposed to policy expansion (e.g. daily car usage, renting, residing in an older house). However, for those with high levels of trust, the negative effect of increased exposure is minimal, decreasing from 2.75 to 2.4 (a 9 per cent change). Similarly, the results hold for political trust, albeit to a lesser extent. While our control variables – encompassing respondents’ political interest, climate concern, political ideology and age – all demonstrate high significance and align with expected trends, we find that neither gender nor income (after accounting for policy vulnerability) influences support.

Comparative Analysis: Hungary

To (externally) validate our results of the effect of energy transition exposure on support for ambitious climate policy, we compare our results from Switzerland and Germany to Hungary. While the former countries have embraced energy transition politically for over 20 years, Hungary has just started on its path to carbon neutrality. Hungary has an energy strategy in place for 2030, but this mostly consists of aspirational goals. The right-wing Fidesz government is against the introduction of a carbon tax, although opposition parties have supported the idea of carbon pricing. Hence, in the wake of the 2022 election (our survey was in the field two weeks before the April elections) we could not ask our Hungarian respondents about a carbon tax increase, but asked their opinions on a possible introduction of such tax. To make the results comparable across countries, we transformed the answers to a binary variable showing whether a respondent supports (1) or opposes (0) the increase (introduction of the) carbon tax (hence excluding the middle category in Hungary). Table 2 shows the models for generalized (Model 1, 3, 5) and political (Model 2, 4, 6) trust in Germany, Switzerland and Hungary. We indeed see that none of our exposure variables influence people’s attitudes towards the introduction of the carbon tax in Hungary where the population is (yet) the least exposed to renewables. Thus, while the German and Swiss results are very similar in direction and significance, the results in Hungary do not show a similar pattern. With respect to our trust measures, we can see that in Hungary, too, generalized trust is a robust predictor of climate policy expansion. Trust in political institutions, however, is not significantly correlated with a more supportive attitude. This again points to the importance of generalized trust as a predictor of collective good provision (Nannestad 2008). Our control variables exhibit a similar correlation to the outcome as in Germany and Switzerland, with two notable differences: in Hungary, women and individuals with higher incomes exhibit a higher rate of support for the introduction of a carbon tax compared to men and those with lower incomes. Detailed models for each individual country, including interactions, are provided in the supplementary materials.

Table 2. Comparative influence of subjective exposure and policy vulnerability on the support of carbon tax increase (D, CH) /introduction (HUN) (Logit models by country).

	Germany		Switzerland		Hungary
DV: support for increase or introduction of the CO2 tax YES/NO	Model 1 Generalized trust	Model 2 Political trust	Model 3 Generalized trust	Model 4 Political trust	Model 5 Generalized trust	Model 6 Political trust
Subjective exposure	0.51 **	0.55 ***	0.87 ***	0.88 ***	0.25	0.27
	(0.16)	(0.17)	(0.25)	(0.25)	(0.26)	(0.26)
Policy vulnerability	−0.20 ***	−0.20 ***	−0.24 ***	−0.23 ***	−0.02	−0.03
	(0.03)	(0.03)	(0.03)	(0.03)	(0.04)	(0.04)
Generalized trust	0.35 ***		0.29 ***		0.32 ***
	(0.04)		(0.05)		(0.06)
Political trust		0.19 ***		0.21 ***		0.03
		(0.02)		(0.02)		(0.03)
Ideology	−0.05 *	−0.06 *	−0.16 ***	−0.17 ***	−0.23 ***	−0.24 ***
	(0.02)	(0.02)	(0.03)	(0.03)	(0.03)	(0.03)
Political interest	0.14 *	0.16 **	0.23 ***	0.18 **	0.38 ***	0.40 ***
	(0.06)	(0.06)	(0.07)	(0.07)	(0.08)	(0.08)
Climate concern	1.02 ***	0.92 ***	1.07 ***	1.04 ***	0.72 ***	0.74 ***
	(0.06)	(0.06)	(0.08)	(0.08)	(0.08)	(0.08)
Income	0.02	−0.01	−0.03	−0.04	0.16 *	0.16 *
	(0.04)	(0.04)	(0.04)	(0.04)	(0.07)	(0.07)
Age	−0.02 ***	−0.02 ***	−0.01 **	−0.01 **	−0.00	0.00
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Gender	−0.12	−0.12	−0.10	−0.05	−0.30 *	−0.29 *
	(0.09)	(0.09)	(0.10)	(0.11)	(0.14)	(0.14)
(Intercept)	−2.70 ***	−2.31 ***	−2.57 ***	−2.71 ***	−1.95 ***	−1.32 **
	(0.37)	(0.37)	(0.44)	(0.44)	(0.51)	(0.49)
BIC	3,516.43	3,456.51	2,458.26	2,397.54	1,460.26	1,477.38
Log likelihood	−1,718.23	−1,688.28	−1,190.87	−1,160.54	−694.38	−702.96
Deviance	3,436.47	3,376.55	2,381.73	2,321.07	1,388.76	1,405.92
Num. obs.	2,971	2,968	2,106	2,093	1,274	1,269

* p < 0.05, ** p < 0.01, *** p < 0.001

Discussion and Conclusion

In this paper, we explored the link between people’s subjective exposure to energy transition, their policy vulnerability and their support for more ambitious climate change policy. Moreover, we questioned whether this relationship was conditional on a person’s generalized and political trust.

From our main analysis, we first find that subjective exposure to energy transition does indeed matter for people’s climate policy preferences. Both measures – subjective exposure to renewable energy infrastructure and individual vulnerability with respect to more ambitious climate policies – have a clear impact on people’s attitudes towards climate policy expansion. To the best of our knowledge, we are the first to explicitly consider measures of subjective exposure to energy transition and their impact on political attitudes. Our argument is that we need to go beyond the standard demographic variables and consider how existing infrastructural changes to people’s communities and local areas may have affected their support for more ambitious climate policy. The good news from our results is that those who are subjectively exposed to past energy transition are actually on average more supportive of climate policy expansion. The bad news, however, might be that the positive effect is mostly driven by exposure to solar panels and parks, while for the exposure to wind turbines, the effect is insignificant. Overall, while energy transition infrastructure might meet opposition locally, our findings show that exposure to it does not necessarily influence people’s general climate policy support.

Second, with respect to the influence of individual policy vulnerability, we corroborate theoretical accounts centering around the importance of distributional effects for climate policy support (Aklin and Mildenberger 2020; Colgan et al. 2021). Based on our empirical evidence, distributional considerations arising out of policy vulnerability are significantly linked to less support for an increase in the CO₂ tax. In our study, we used the CO₂ tax as a climate policy instrument that has very clear distributional consequences, but also a very clear link to effective emissions reductions. In some ways, this could be considered an upper bound of estimates with respect to the link to an individual’s policy vulnerability. Other instruments that have less direct effects on people’s pocketbooks, such as subsidies for renewable energy or for the uptake of electric vehicles, may generate different results with respect to policy support. Future research might evaluate how the relationships between exposure, vulnerability and trust play out for different policy instruments.

Third, we add to a broad literature that has consistently pointed to the importance of trust for individual (climate policy) attitudes (Hammar and Jagers Sverker 2006; Delhey et al. 2011; Fairbrother 2019; Fairbrother et al. 2019; Davidovic and Harring 2020; Umit and Schaffer 2020). In our contribution, we develop arguments on how individual trust conditions the (proposed negative) relationship between exposure and support for climate policy expansion. Our evidence shows that trust influences the relationship between exposure and policy support. We find that trusting individuals seem generally more likely to accept long-term governmental decisions (e.g. on energy transition) and support their implementations, even if they jeopardize their self-interest. Especially high generalized trust (but also political trust) is associated with a more positive (less negative) influence on climate policy support levels for exposed individuals. Overall, higher generalized and political trust can help to improve the positive and lessen the negative effects of subjective exposure and policy vulnerability on the support for policy expansion.

We validate our results obtained from our two main country cases (Germany and Switzerland), by considering the influence of exposure and vulnerability on climate policy expansion and the conditioning effect of trust in a third country (Hungary) that is comparatively less advanced on its path towards energy transition. We find corroborating evidence indicating that in Hungary, where people remain less exposed to renewable infrastructure, our exposure variables are not linked to an individual’s policy support for the introduction of a carbon tax. Also in Hungary, generalized trust is associated with increased support for the tax. To gauge the generalizability of our results, future studies might want to extend our country selection to include more countries that have so far been comparatively less committed to energy transition or that exhibit different climatic characteristics and levels of exposure to energy transition.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Supplementary Material

Supplementary data for this article can be accessed at https://doi.org/10.1080/13876988.2024.2331166

Additional information

Funding

The research in this paper is part of the project ‘Beyond Policy Adoption: Implications of Energy Policy on Parties, Publics, and Individuals’, funded by the Swiss National Science Foundation (PYAPP1–173642/1)..

Notes on contributors

Lena Maria Schaffer

Lena Maria Schaffer is Professor of Inter- and Transnational Relations at the University of Luzern in Switzerland. Her main research interests are in the area of international and comparative climate and energy policy, and individual preferences regarding climate change and globalization.

Zsuzsanna B. Magyar

Zsuzsanna B. Magyar is a Post-doctoral Research Fellow at the Institut Barcelona d’Estudis Internacionals Barcelona, Spain. Her main reseach interests are party politics, party systems, comparative political institutions, historical political economy and comparative public policy.

Notes

1. For a recent comprehensive review of public opinion research on energy infrastructure, see Carley et al. (2020).

2. In the case of Germany, early movers could even attain considerable remuneration due to generous feed-in tariffs. There, the generous remuneration for providers of renewable energy, however, have also led to negative repercussions of energy transition as energy prices have increased (Bardt and Niehues 2016; Renn and Marshall Jonathan 2016).

3. In more urban or sunny regions, solar panels are a common sight for the people living there.

4. The samples are representative of age, education, and rural–urban classification in Switzerland and Germany. In Hungary, the sample is representative of age and rural–urban classifications.

5. The original questions are provided in online Appendix C1.

6. We do not directly compare the two variables in this study, but use a binary variable (oppose/support) for the ancillary comparative analysis. First, the introduction (Hungary) and a subsequent increase (Germany, Switzerland) of a carbon tax make people evaluate different concepts in our view. Second, in Hungary we have a middle category (do not support, do not oppose). In Hungary we create the binary variable by using the answers support and strongly support as the positive variable and oppose or strongly oppose as a negative variable. As a robustness check we use middle categories and do not knows in the estimation but this does not change the results, see Table B8 in online Appendix B.

7. For Germany, online Appendix Figure A1b additionally depicts the regional variation of our subjective exposure variable.

8. Recently, the importance of the distributional divide between home owners and tenants in progressing energy transition has generated some scholarly interest (Jansma et al. 2022; Schaffer 2023; Voeten 2023). Home owners can more easily participate in and make their property fit for the energy transition (buying solar panels, changing their heating system, electrifying their homes) to avert rising prices due to carbon taxes, while tenants often have to cope with higher energy prices without being able to do much against this. Their landlords would have to do it, but as energy costs are mostly borne by the tenants themselves they have little incentive to do so.

9. To account for the possibility that trust and political trust could be highly correlated, and thus the two specifications measure the same thing, we also ran the general trust models controlling for political trust. We find that the two types of trusts have an independent association with support for carbon taxes. See the results in online Appendix B, Table B7.

10. We do this, first, to account for the unobserved heterogeneity between the cases. For this purpose, the models contain country-region dummies. The mixed-effect model combines the fixed-effects and mixed-effects models in a hierarchical way. With the fixed effects we control for the unmeasured characteristics within each individual region. However, we also believe that the provinces are related to each other and the concern for environmental issues became such a global phenomenon that we cannot treat the cases as isolated, and thus we want to account for the possibility that the people in each province are exposed to similar news and trends. Specifically, we believe that people with different levels of trust in the different regions might relate to government policies similarly. We allow for this possibility with the random effects.

11. In online Appendix B we show the same models with fixed effects (Table B2–B5) and models with a binary dependent variable (support, do not support) (Table B6). Furthermore, we show that the results hold in Hungary as well (Table B4, Figure B4).

12. We also ran the regressions when we only included the exposure to solar panels and its interactions with trusts. We find that indeed exposure to solar panels drives our results. We report these results in online Appendix B, Table B6.

13. We gathered objective measures of exposure to renewable energy sources in Switzerland and Germany by calculating financial flows associated with EEG renumerations in Germany and Kostendeckende Einspeisevergütung (KEV) renumerations in Switzerland. As the financial flows depend on the renewable capacity within the area, it can serve as a good proxy of how locally exposed people are. Including these measures as controls did not alter the results. These findings are presented in online Appendix B, Table B1.